Machine Learning for Data Mining

Implementations of propensity scores

To see how we can use propensity scores to our advantage, follow these steps:

1. Get your dataset on the canvas and connect it to a Partition node, dividing the dataset into two parts.
2. Connect the Partition node to a Chaid model from the Modeling tab. You could use any model here, but let's use this as it will be used in our next example as well.

3. Chaid will build a decision tree model. To edit, go to the Model Options tab, where there is a section that asks for Propensity Scores. There are two types; a raw propensity score is for the training dataset and the adjusted propensity score is for the testing or validation dataset. We will select the raw propensity score for now:

4. Click on Run. Connect the ...